Cite
Notes
Only stored in your browser.
Attribution
Trainable Dynamic Mask Sparse Attention
arXiv 2025
Concise Reasoning, Big Gains: Pruning Long Reasoning Trace with Difficulty-Aware Prompting
Wonderful Matrices: Combining for a More Efficient and Effective Foundation Model Architecture
arXiv 2024
from 3 papers
Jingze Shi
Yifan Wu
Yuyu Luo
Guang Liu
Jiayi Zhang
Liangdong Wang
Nan Tang
Xiaotian Lin
Yiran Peng